Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-08-02

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:27 xarses joined #fuel
00:58 code-R joined #fuel
01:00 code-R_ joined #fuel
01:48 ilbot3 joined #fuel
01:48 Topic for #fuel is now Fuel 8.0 (Liberty) https://www.fuel-infra.org/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
04:15 yottatsa joined #fuel
04:26 Zer0Byte__ joined #fuel
04:42 Zer0Byte__ joined #fuel
04:52 elo joined #fuel
05:11 yottatsa joined #fuel
05:14 yottatsa_ joined #fuel
05:15 code-R joined #fuel
05:21 code-R_ joined #fuel
05:52 maestropandy joined #fuel
06:00 javeriak joined #fuel
06:16 code-R joined #fuel
06:18 cartik joined #fuel
06:25 vkulanov joined #fuel
06:26 Miouge joined #fuel
06:29 code-R_ joined #fuel
06:47 code-R joined #fuel
07:20 Venkat joined #fuel
07:21 venkat joined #fuel
07:24 TiDjY35 joined #fuel
07:30 gcmacmalloc joined #fuel
07:34 coreping joined #fuel
07:43 aarefiev joined #fuel
07:46 asvechnikov ikutukov, ikalnitsky core-review please revert https://review.openstack.org/#/c/348853/
07:49 javeriak hey folks, where should i pick appropriate package_version in metadata.yml for fuel-plugins? i see some plugins using 2.0.0 and some using 4.0.0
08:08 noshankus joined #fuel
08:15 nokes joined #fuel
08:18 v1k0d3n joined #fuel
08:18 kgalanov joined #fuel
08:21 gbraad joined #fuel
08:28 yassou joined #fuel
08:55 silvio joined #fuel
08:59 skolekonov joined #fuel
09:06 yottatsa joined #fuel
09:23 romcheg jaranovich: ping, around?
09:23 jaranovich romcheg: yep
09:26 Guest54610 hi all, I have problems setting up fuel on esxi, can someone help me a little bit please?  VMs dont get dhcp/pxe ... Will Fuel act as gateway if I only have a vSwitch?
09:26 HeOS joined #fuel
09:28 aderyugin joined #fuel
09:28 romcheg jaranovich: do you have an environment to test my patch that adds timestamps to tasks?
09:30 jaranovich romcheg: i guess i can test your patch on fake env
09:37 romcheg jaranovich: so I've prepared this patch https://review.openstack.org/#/c/349901/
09:38 romcheg it has no tests so I cannot guarantee that it even works
09:38 romcheg however I would appreciate if you took a look at it and tell me if it will resolve your issue
09:38 jaranovich i'll check it on fake env
09:38 ekosareva joined #fuel
09:39 romcheg meanwhile I will be working on tests and fixing the code if it does not work
09:47 javeriak joined #fuel
09:53 romcheg vkulanov, ikalnitsky: could you guys please review this one https://review.openstack.org/#/c/344320/ ?
09:57 jaranovich romcheg: checked your patch and couldn't  get transactions. Posted a comment to the request, please consider it
10:14 vkulanov romcheg, see my comments
10:14 romcheg vkulanov: thanks! I will address them ASAP
11:18 yassine joined #fuel
11:21 asvechnikov ikutukov, ikalnitsky core-review please revert https://review.openstack.org/#/c/348853/
11:24 javeriak joined #fuel
11:55 javeriak joined #fuel
11:56 fritchie joined #fuel
12:00 fritchie joined #fuel
12:32 javeriak joined #fuel
12:39 vkulanov romcheg, can you, please, take a look. Should we should we backport this fix https://bugs.launchpad.net/fuel/+bug/1603440 to mitaka?
12:39 romcheg dpyzhov ^
12:44 dpyzhov @vkulanov
12:44 dpyzhov yep, let's backport
12:51 vkulanov dpyzhov, romcheg ok, thanks
12:53 maestropandy left #fuel
13:09 flerfb0rt1 joined #fuel
13:23 ekosareva joined #fuel
13:40 thiagolib joined #fuel
13:41 ekosareva Colleagues, please review https://review.openstack.org/#/c/341355/
13:49 code-R joined #fuel
13:52 code-R_ joined #fuel
13:53 thumpba joined #fuel
14:01 code-R joined #fuel
14:01 gcmacmalloc joined #fuel
14:02 code-R joined #fuel
14:13 sneha @mwhahaha In reference to my last question, if I make changes in the fuel code and pass my own package/service names then do I need to place my packages on fuel repos or it is fine that they are present on /var/www/.. ?
14:14 mwhahaha You would need to add them to a repo that is configured for the environment
14:15 mwhahaha If you had just used the same package names you wouldn't need to update the params.pp but the repo would still be a requirement
14:17 sneha also if I am changing fuel code; then can I use fuel's netconfig itself? or do I need to skip Fuel's netconfig and create my own task for netconfig ?
14:21 sneha and by repo that is configured for the environment u mean : /var/www/nailgun/plugins/plugin_name/repositories right ?
14:22 mwhahaha so the plugin repository is automatically populated via the plugin
14:22 mwhahaha so you'd have to bundle your package with the plugin
14:23 mwhahaha but if you're doing that, you could just rename the package/service to match the expected upstream names because i think the plugin has a higher priority
14:28 sneha renaming package/service is impossible..our firm wont permit it
14:30 sneha how can i bundle my package with the  plugin?
14:30 mwhahaha the plugin structure includes a repository/ folder
14:31 mwhahaha just drop it in the ubuntu folder
14:31 mwhahaha then when you build it, it'll automatically get put into the plugin repo that is created when you install/enable your plugin
14:39 xarses joined #fuel
14:39 Bomfunk joined #fuel
14:42 sneha earlier I used to install the plugin and manually roll the packages in the ubuntu folder in the plugin repo. Will that be necessary if I have already added it in the repository/folder before installing the plugin ?
14:47 mwhahaha when you put it in the repository/ubuntu folder then build it, it's automatically bundled with the plugin and the installation process of the plugin puts it in the repo for nailgun
14:47 mwhahaha so if it's included in the plugin, you don't have to manually do any repo setup because it'll be done for you.
14:47 mwhahaha every plugin gets a repo configured on the deployed nodes, it's just usually empty
14:50 sneha but if i manually roll the packages in the ubuntu folder in the plugin repo will my approach with changes in the params.pp still work ?
14:55 mwhahaha you'd still have to do the params.pp if you want it to use the other packages
14:56 mwhahaha it's two things, 1) make the packages available via a repo and 2) update the code so it actually uses those package/service names
14:56 mwhahaha if you wish to distribute this plugin, you'd probably want to just implement your own netconfig task that replaces the other and not do the params.pp update
15:14 fritchie joined #fuel
15:19 fritchie joined #fuel
15:31 thumpba joined #fuel
15:32 thumpba joined #fuel
15:41 code-R joined #fuel
15:43 fritchie joined #fuel
15:52 venkat joined #fuel
15:59 krypto joined #fuel
15:59 jackdastripper mwhahaha: I tried using yesterday's 10.0 ISO and it also fails when selecting L3 HA. Here's the bug report I just created. https://bugs.launchpad.net/fuel/+bug/1609013
16:00 mwhahaha K I'll look at it in bit
16:00 jackdastripper It seems that Mysql is failing to install or startup correctly. Seems to happen on 1 of the 2 controller nodes.
16:02 Sesso joined #fuel
16:05 code-R_ joined #fuel
16:13 javeriak joined #fuel
16:16 fritchie joined #fuel
16:19 javeriak_ joined #fuel
16:21 Miouge joined #fuel
16:27 elopez joined #fuel
16:44 mwhahaha jackdastripper: were you only deploying 2 controllers nodes? you need 3
16:45 thumpba joined #fuel
16:52 mwhahaha fuel-library folks please review https://review.openstack.org/#/c/349721/ as it's a swarm blocker, holser_ alex_didenko aglarendil mattymo
16:58 brig-man How does fuel in discovery handle multi-port devices?
16:59 mwhahaha what do you mean multi-port devices?
17:00 brig-man SAS device - two port.  Two paths to same device.
17:03 brig-man I have a system with 12 SAS devices and discovery is trying to use it as 24 devices.
17:05 Sketch you can pick what devices to use when you discover nodes in fuel
17:06 Sketch from Configure Disks, though it wasn't immediately obvious to me that you could change things there
17:06 Sketch there may also be a way to template it like you can for network devices?
17:06 mwhahaha oh multipath
17:06 mwhahaha i think we've added support in 9 for that, not completely sure
17:06 fritchie joined #fuel
17:12 thumpba joined #fuel
17:15 HeOS joined #fuel
17:22 thumpba_ joined #fuel
17:36 maestropandy joined #fuel
17:37 fritchie joined #fuel
17:38 code-R joined #fuel
17:42 brig-man Is there a config option to add the IPMI IP address for the nodes?
17:42 fritchie joined #fuel
17:43 maestropandy left #fuel
17:46 fritchie joined #fuel
17:59 mwhahaha no, we don't have any ipmi integration for the nodes within fuel as far as i k now
18:15 Miouge joined #fuel
18:17 gcmacmalloc joined #fuel
18:30 Miouge joined #fuel
18:35 jackdastripper mwhahaha: Yes only two controllers. I'm tight on disk space until I get the first compute/ceph node online. I thought three was a suggested minimum, not a hard minimum. I understand you need two nodes to do HA  routers.
18:35 mwhahaha yea but in order to do ha for the controllers you need 3
18:36 mwhahaha the issue is with the cluster formation around the controllers you need 1,3,5,etc
18:37 jackdastripper okay, I would think it shouldn't let you deploy just two. I should be able to get three running with the virtual drives thin provisioned.
18:38 mwhahaha there are cases where 2 might be ok, but if you're so concerned about the l3 ha then you really should have 3 proper controllers
18:45 jackdastripper mwhahaha: I guess my question is only having two controllers the source of the deployment failure or is that just the best practice for a production deployment?
18:45 mwhahaha its the source of your deployment failure and a best practice :D
18:46 mwhahaha so either go with 1 and not have HA (for controllers or l3) or create 3 controllers and turn on the HA l3 items
18:47 mwhahaha the latter would be a proper HA configuration
18:55 fritchie joined #fuel
18:58 maestropandy1 joined #fuel
18:58 maestropandy1 left #fuel
19:04 Miouge joined #fuel
19:10 Miouge joined #fuel
19:19 maestropandy joined #fuel
19:19 maestropandy left #fuel
19:32 fritchie joined #fuel
19:33 maestropandy1 joined #fuel
19:33 maestropandy1 left #fuel
19:36 degorenko joined #fuel
19:38 iberezovskiy joined #fuel
19:42 maestropandy joined #fuel
19:47 maestropandy left #fuel
19:54 javeriak joined #fuel
19:58 Miouge joined #fuel
19:58 fritchie joined #fuel
20:00 fritchie joined #fuel
20:11 brig-man Can I get some help troubleshooting network validation error that seems to be software?
20:12 brig-man message back nodes not avaliable via mcollective, yet mco ping shows all nodes.
20:37 fritchie joined #fuel
21:16 brig-man If I have upgraded the fuel node, do I need to rebuild my bootstrap images?
21:18 xarses brig-man: you should at least re-boot the discovered nodes, it's probably related to the error you are seeing
21:18 xarses if you rebuild the bootstrap image, it will get any new os updates, and if any packaged where added vs the prior version
21:19 code-R joined #fuel
21:22 akscram brig-man: In addition, after an upgrade of the fuel master node you need to restart mcollective on already provisioned or deployed nodes
21:23 brig-man xarses: I have rebooted the nodes after the fuel node update.
21:23 akscram because some of them could lost connection to rabbit and they should not appear in mco ping
21:24 akscram brig-man: Oh, could you elaborate what do you mean 'upgraded the fuel node'?
21:25 brig-man dockerctl backup; yum update; docker load -i /var/www/nailgun/docker/images/fuel-images.tar; dockerctl destroy all; dockerctl start all
21:26 brig-man I have not slave fuel nodes.
21:26 brig-man no slave fuel nodes
21:26 brig-man The nodes in question show up in networking but I must manually start mcollective on them.
21:27 brig-man ping/ssh
21:31 code-R_ joined #fuel
21:42 Egyptian[Home] joined #fuel
21:50 brig-man The four nodes that are failing were previously configured.  In discovery, these are not starting mcollective.  I reset the environment hoping to get out of this failed state but it seems to continue to spiral to new problems.
21:56 brig-man OK.  all the failing nodes show this in the agent log: Can't get data from lshw. Reason: No such file or directory
21:57 brig-man Red herring.  non-error nodes show the same message.
21:59 Egyptian[Home] joined #fuel
22:00 fritchie joined #fuel
22:21 elo joined #fuel
22:39 Egyptian[Home] joined #fuel
22:59 Egyptian[Home] joined #fuel
23:26 brig-man on Fuel security setup, can you have multiple address for ssh?
23:32 johnavp1989 joined #fuel
23:55 code-R joined #fuel
23:57 code-R_ joined #fuel

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary